CN107833220A - Fabric defect detection method based on depth convolutional neural networks and vision significance - Google Patents

Fabric defect detection method based on depth convolutional neural networks and vision significance Download PDF

Info

Publication number
CN107833220A
CN107833220A CN201711212830.0A CN201711212830A CN107833220A CN 107833220 A CN107833220 A CN 107833220A CN 201711212830 A CN201711212830 A CN 201711212830A CN 107833220 A CN107833220 A CN 107833220A
Authority
CN
China
Prior art keywords
mrow
defect
msub
pixel
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201711212830.0A
Other languages
Chinese (zh)
Other versions
CN107833220B (en
Inventor
李庆武
邢俊
马云鹏
周亚琴
吴晨辉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Changzhou Campus of Hohai University
Original Assignee
Changzhou Campus of Hohai University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Changzhou Campus of Hohai University filed Critical Changzhou Campus of Hohai University
Priority to CN201711212830.0A priority Critical patent/CN107833220B/en
Priority to PCT/CN2017/116837 priority patent/WO2019104767A1/en
Publication of CN107833220A publication Critical patent/CN107833220A/en
Application granted granted Critical
Publication of CN107833220B publication Critical patent/CN107833220B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G06T7/0008—Industrial image inspection checking presence/absence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/155—Segmentation; Edge detection involving morphological operators

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Quality & Reliability (AREA)
  • Health & Medical Sciences (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

The invention discloses a kind of fabric defect detection method based on depth convolutional neural networks and vision significance, belong to technical field of image processing.Include defect area locating module and defect semantic segmentation module.Defect area locating module is merged using two deep learning models of local convolutional neural networks and global convolutional neural networks, is automatically extracted the advanced features of fabric defect and is acted on defect image, obtains being accurately positioned for defect area.Defect semantic segmentation module utilizes the positioning result of defect area, with reference to the super-pixel image partition method of view-based access control model conspicuousness, obtains defect priori foreground point and defect target is precisely split, finally realize the detection of defect.The present invention utilizes the fabric defect positioning network of more deep learnings fusion and the fabric defect segmentation network of improved vision significance, it is good to the adaptability of textile image, precision is high, and effectively the defects of complex background and textile image under noise jamming can be detected.

Description

Fabric defect detection method based on depth convolutional neural networks and vision significance
Technical field
The present invention relates to the field of visual inspection in image procossing, more particularly to it is a kind of based on depth convolutional neural networks with The fabric defect detection method of vision significance.
Background technology
With the rapid development of textile industry, control of the people to fabric cloth quality is also increasingly stricter, and fabric defect Point is typically the key factor for influenceing cloth quality.Traditional fabric defect detection method is to be based on hand dipping and human eye mostly Observe to complete, there is significant limitation in actual applications, as subjectivity is strong, testing result uniformity is poor, it is impossible to very accurate True realization is to tiny defect, the complete detection etc. of aberration unobvious defect.At present, existing automation fabric defect detection is calculated Method is broadly divided into three classes：(1) method of method, (3) based on model of Statistics-Based Method, (2) based on spectrum analysis.Based on mould The method of type is obtained the texture information of textile image and reconstructs zero defect image, passed through using defect image structure decomposition model Compare input picture and reconstruct the difference between zero defect image to position defect.The usual accuracy of such method is not high, and With sizable computation complexity.Statistics-Based Method is carried using Fourier transform, Gabor filter and wavelet transformation The frequency domain characteristic of defect image is taken, its algorithm performance is heavily dependent on wave filter species and fabric selected by algorithm The background of image.Statistics-Based Method is united by the methods of local binary patterns, gray level co-occurrence matrixes and statistics with histogram Texture and the different qualities of defect are counted, can effectively detect fabric defect, but the difference of fabric background pattern and defect shape Such method is had a great influence.
The studies above can be positioned and split to the fabric defect under uniform background and solid background exactly, but right It is bad in the defects of background complexity, the irregular textile image of pattern Detection results.
The content of the invention
The technical problems to be solved by the invention are：A kind of fabric defect detection method is provided, with realize it is complicated to background, The defects of irregular textile image of pattern is effectively detected.
To achieve these goals, the present invention takes following technical scheme.
A kind of fabric defect detection method based on depth convolutional neural networks and vision significance, comprises the following steps：
(1) fabric defect training dataset is chosen, gray processing processing is carried out to the image that data are concentrated, then carries out size Normalized；
(2) will be inputted by the pretreated fabric defect training dataset of step (1) to defect area locating module, institute Defect area locating module is stated respectively to carry out fabric data set using global neural network model and Local neural network model Training, overall situation and partial situation's advanced features of fabric defect are extracted, obtain the minimum model of an error；
(3) image to be tested is subjected to gray processing and normalized, is then separately input into global neural network model With Local neural network model；Each pixel that global neural network model is responsible for treating in test image is predicted, defeated Go out the probability vector that each pixel belongs to defect area；Local neural network model is not responsible to be carried out in advance to each pixel Survey, but treat the initial alignment that test image carries out defect area, obtain the bounding box of some defect areas, the bounding box It is defect candidate region；
(4) prediction result and Local neural network model of each pixel are exported using global neural network model The defects of region bounding box, pass through the multi-model fusion method of structure, obtain and combine global neural network model, local nerve The defects of two models of network model area score, defect area is rejected according to the score；
(5) using SLIC super-pixel segmentations algorithm by defecton image region segmentation into several different super-pixel areas Domain, a node is regarded as in each super-pixel region, then closed using the region contrast between super-pixel node, locus System, the notable function of priori part heatmap information architecture super-pixel nodes, and according to notable function calculating input image Priori notable figure；
(6) due to priori notable figure generally exist salient region saliency value is inconsistent, background area can not press down very well The problems such as processed, therefore priori notable figure is split using auto-thresholding algorithm, it is then based on morphology opening and closing fortune Calculation post-processes to image after segmentation, removes image cavity and some scatterplots, finally detects the defects of textile image.
The gain effect that the present invention is reached：
The present invention is done for complex background with noise it can be seen from the technical scheme provided by the example of the invention described above Textile image under disturbing, propose a kind of based on the positioning of the fabric defect of depth convolutional neural networks and vision significance and detection side Method, fusion is thought the Primary Location of defect using classification of the global neutral net to pixel and Local neural network to obtain precisely The defects of anchor window, be then based on improved vision significance method and the defects of anchor window split.This method Artificial setup parameter and structure reference picture are not needed, robustness is good, can be accurately detected the defects of textile image, real Shi Xingqiang, Practical Project demand can be met, be with a wide range of applications.
Brief description of the drawings
Fig. 1 is the flow of the fabric defect detection method based on depth convolutional neural networks and vision significance of the present invention Figure；
Fig. 2 is fabric defect detection model schematic diagram；
Fig. 3 is global convolutional neural networks model schematic；
Fig. 4 is local convolutional neural networks model schematic；
Fig. 5 is based on parted pattern schematic diagram the defects of improving vision significance.
Embodiment：
Further detailed description is done to the embodiment of the present invention below in conjunction with the accompanying drawings.
As depicted in figs. 1 and 2, the fabric defect detection side of the invention based on depth convolutional neural networks and vision significance Method, including defect area positioning network model and defect Segmentation network model.Defect location network model utilizes global nerve net Network model blends with Local neural network model, there is provided defect accurate positional information in textile image.Defect Segmentation net Network model is split to defect area using super-pixel and vision significance content, is extracted defect target.Including following step Suddenly：
(1) fabric defect training dataset is chosen, gray processing processing is carried out to the image that data are concentrated, then carries out size Normalized；
(2) will be inputted by the pretreated fabric defect training dataset of step (1) to defect area locating module, institute Defect area locating module is stated respectively to carry out fabric data set using global neural network model and Local neural network model Training, overall situation and partial situation's advanced features of fabric defect are extracted, obtain the minimum model of an error；
(3) image to be tested is subjected to gray processing and normalized, is then separately input into global neural network model With Local neural network model；Each pixel that global neural network model is responsible for treating in test image is predicted, defeated Go out the probability vector that each pixel belongs to defect area；Local neural network model is not responsible to be carried out in advance to each pixel Survey, but treat the initial alignment that test image carries out defect area, obtain the bounding box of some defect areas, the bounding box It is defect region that may be present；
(4) prediction result and Local neural network model of each pixel are exported using global neural network model The defects of region bounding box, pass through the multi-model fusion method of structure, obtain and combine global neural network model, local nerve The defects of two models of network model area score, defect area is rejected according to the score；
(5) using SLIC super-pixel segmentations algorithm by defecton image region segmentation into several different super-pixel areas Domain, a node is regarded as in each super-pixel region, then closed using the region contrast between super-pixel node, locus System, the notable function of priori part heatmap information architecture super-pixel nodes, and according to notable function calculating input image Priori notable figure；
(6) due to priori notable figure generally exist salient region saliency value is inconsistent, background area can not press down very well The problems such as processed, therefore priori notable figure is split using auto-thresholding algorithm, it is then based on morphology opening and closing fortune Calculation post-processes to image after segmentation, removes image cavity and some scatterplots, finally detects the defects of textile image.
In step (1), brightness Y and tri- color components of R, G, B are established using the variation relation of RGB and YUV color spaces Relation, i.e. Y=0.11B+0.59G+0.3R, gray processing is carried out to fabric defect image data set, and utilizes bi-cubic interpolation The training image is normalized to set pixel size, such as 400 × 400 pixel sizes by method.
In step (2), when training global neural network model, the input for training global neutral net is fabric defect image Data set and fabric defect mark index map, the global characteristics of textile image, wherein convolution kernel size are extracted using convolution operation 3 × 3 sizes are taken, as shown in Figure 3.
The model parameter of global neutral net is：
First layer is image input layer, and image input layer size is consistent with training picture size, is taken herein 400 × 400 big It is small；Training picture refers in step (1) that the picture that fabric defect training data is concentrated, its size is 400 × 400；
There are l layer hidden layers centre, is replaced by convolutional layer with pond layer and connected and composed；L can be 8 layers；
It is finally some warp laminations, because convolutional layer causes the spy of 10 × 10ppi~200 × 200ppi resolution sizes Sign figure recovers to artwork 400 × 400ppi sizes, and each pixel belongs to the probability of defect point in output image, obtains one Heatmap schemes.Ppi implication refers to the pixel number that per inch image is included, and artwork refers to 400 × 400 input picture, That is 400 × 400ppi resolution ratio；
In step (2), when training local neural network model, the input of Local neural network for fabric defect data set and The coordinate of fabric defect in the picture, extracts the local feature of textile image using convolution operation, wherein convolution kernel size with it is complete Convolution kernel in office's neural network model is in the same size, is 3 × 3 sizes, as shown in Figure 4.
The model parameter of Local neural network is：
First layer is training image input layer, and the size of image input layer, which is normalized into, to be sized, such as 300 × 300 Size；There are k layer hidden layers centre, is replaced by convolutional layer with pond layer and connected and composed；Finally full articulamentum, output defective locations with Classification information.
In step (2), global neutral net and Local neural network model are carried out respectively using n fabric defect images K is respectively trained with Local neural network model in training, global neutral net1Secondary and k2After secondary, model error convergence, now obtain Optimal models weight.
In step (3), network model is set using the optimal models weight trained, test fabric image is inputted respectively Into global neural network model and Local neural network model, the positional information of heatmap figures and defect is exported respectively.
In step (3), the heatmap figures of global neural network model output are a kind of index maps, and each numerical value represents one Kind of color, color represent that defect possibility is higher, and it is lower to represent defect possibility closer to blueness for color closer to red.
In step (3), the positional information of Local neural network output, its data format is [xmin,ymin,xmax,ymax, Label], wherein xmin, ymin, xmax, ymaxIt is the upper left corner and the bottom right angular coordinate of defect bounding box respectively, label is defect institute Belong to species.
In step (4), following Combination score function is constructed to calculate defect score Pscore(m)：
Wherein, i, j represent the abscissa and ordinate of pixel in the picture respectively, and A (m) represents Local neural network model M-th of detection window in SSD, A (m) .conf represent the score of m-th of testing result in Local neural network model SSD, S The size of A (m) window is represented, B (i, j) represents pixel value of pixel of the coordinate as (i, j) in heatmap.
In step (4), threshold value T is compared in setting, is realized by comparing the defects of calculating score with threshold size；If The defects of calculating score is less than given threshold T, then it is assumed that the testing result is false retrieval, is directly filtered out；If defect obtains Divide and be more than given threshold T, then retain the defect area.
In step (5), fabric defect region subgraph is divided into K super-pixel using SLIC super-pixel segmentations algorithm Block, as shown in figure 5, and using the heatmap figure extractions priori foreground point obtained in global neural network model, before priori Sight spot and region contrast, regional space relation structure super-pixel significance function gamma (Pi)：
Wherein, i1, j1Super-pixel node serial number is represented respectively,For i-th1Individual super-pixel node,For super-pixel nodeThe normalization Euclidean distance of super-pixel node where to priori foreground point；For super-pixel nodeWith super-pixel NodeBetween normalization Euclidean distance；For super-pixel nodeCorresponding region Lab color spaces average value, For super-pixel nodeCorresponding region Lab color spaces average value,The super-pixel region where priori foreground point In the average value of Lab color spaces, K is super-pixel number, and α is regulatory factor.
In step (5), during using heatmap figure extraction priori foreground points in global neural network model, basis first Defecton image coordinate extracts local heatmap figures in heatmap figures, then using N × N cover module in part Slided on heatmap, N is odd number, travels through the position that whole Zhang Suoshu parts heatmap images obtain maximum output value, maximum defeated Go out to be worth the priori prospect point coordinates of the i.e. defect in position；Cover module weight matrix isi2,j2Generation Coordinate of the table pixel in cover module, its specific formula are
Wherein δ () is impulse function, and u () is jump function, and N is template size.Obtained in step (4) Defect area, defect subgraph exactly pluck out each defect area from textile image, a sheet by a sheet image are formed, due to these Image is all the part in original fabrics image, so referred to as subgraph.
In step (5), the priori significance of each super-pixel point in textile image is calculated using significance function, and will Pixel value of the priori saliency value as super-pixel point, build priori notable figure.
In step (6), priori notable figure is split using adaptive threshold OTSU algorithms, the defects of extraction image Target, adaptive threshold choose formula and are：
Wherein α1, α2Respectively the threshold percentage factor one and the threshold percentage factor two, b1, b2Respectively threshold value shift factor One and threshold value shift factor two, IpriorFor the pixel value of priori foreground point.
The present invention is disclosed with preferred embodiment above, so it is not intended to limiting the invention, all to use equivalent substitution Or the technical scheme that equivalent transformation mode is obtained, it is within the scope of the present invention.

Claims (14)

  1. A kind of 1. fabric defect detection method based on depth convolutional neural networks and vision significance, it is characterised in that including Following steps：
    (1) fabric defect training dataset is chosen, gray processing processing is carried out to the image that data are concentrated, then carries out size normalizing Change is handled；
    (2) will be inputted by the pretreated fabric defect training dataset of step (1) to defect area locating module, it is described to lack Zone location module is fallen into respectively to be trained fabric data set with Local neural network model using global neural network model, Overall situation and partial situation's advanced features of fabric defect are extracted, obtain the minimum model of an error；
    (3) image to be tested is subjected to gray processing and normalized, is then separately input into global neural network model and office Portion's neural network model；Each pixel that global neural network model is responsible for treating in test image is predicted, and output is every Individual pixel belongs to the probability vector of defect area；Local neural network model treats test image and carries out the initial of defect area Positioning, obtains the bounding box of defect area, the bounding box is defect candidate region；
    (4) lacked using global neural network model to what the prediction result and Local neural network model of each pixel exported The bounding box in region is fallen into, by the multi-model fusion method of structure, obtains and combines global neural network model, Local neural network The defects of two models of model area score, defect area is rejected according to the score；
    (5) using SLIC super-pixel segmentations algorithm by defecton image region segmentation into several different super-pixel regions, A node is regarded in each super-pixel region as, then utilizes the region contrast between super-pixel node, spatial relation, elder generation The notable function of local heatmap information architectures super-pixel node is tested, and according to the priori of notable function calculating input image Notable figure；
    (6) priori notable figure is split using auto-thresholding algorithm, is then based on morphology opening and closing operation to dividing Cut rear image to be post-processed, remove image cavity and scatterplot, finally detect the defects of textile image.
  2. 2. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (1), brightness Y and R, G, B tri- colors point are established using RGB and YUV color spaces variation relation The relation of amount, i.e. Y=0.11B+0.59G+0.3R, gray processing is carried out to fabric defect image data set, and inserted using double cubes The training image is normalized to set pixel size by value method.
  3. 3. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (2), when training global neural network model, the input for training global neutral net is fabric defect Image data set and fabric defect mark index map, the global characteristics of textile image, global nerve net are extracted using convolution operation Network model parameter is：
    First layer is image input layer, and image input layer size is consistent with training picture size；
    There are l layer hidden layers centre, is replaced by convolutional layer with pond layer and connected and composed；
    It is finally some warp laminations, because convolutional layer causes the characteristic pattern of 10 × 10ppi~200 × 200ppi resolution sizes Recover to artwork size, each pixel belongs to the probability of defect point in output image, obtains a heatmap figure.
  4. 4. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (2), when training local neural network model, the input of Local neural network is fabric defect data The coordinate of collection and fabric defect in the picture, the local feature of textile image, wherein convolution kernel size are extracted using convolution operation In the same size with the convolution kernel in global neural network model, the model parameter of Local neural network is：
    First layer is training image input layer, and the size of image input layer, which is normalized into, to be sized；There are k layer hidden layers centre, Replaced by convolutional layer with pond layer and connected and composed；It is finally full articulamentum, exports defective locations and classification information.
  5. 5. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (2), using n fabric defect images respectively to global neutral net and Local neural network model It is trained, k is respectively trained with Local neural network model in global neutral net1Secondary and k2After secondary, model error convergence, now Obtain optimal models weight.
  6. 6. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (3), network model is set using the optimal models weight trained, test fabric image is distinguished Input exports the positional information of heatmap figures and defect respectively into global neural network model and Local neural network model.
  7. 7. the fabric defect detection method according to claim 6 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (3), the heatmap figures of global neural network model output are a kind of index maps, and each numerical value represents A kind of color, color represent that defect possibility is higher, and it is lower to represent defect possibility closer to blueness for color closer to red.
  8. 8. the fabric defect detection method according to claim 7 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (3), the positional information of Local neural network output, its data format is [xmin,ymin,xmax,ymax, Label], wherein xmin, ymin, xmax, ymaxIt is the upper left corner and the bottom right angular coordinate of defect bounding box respectively, label is defect institute Belong to species.
  9. 9. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (4), following Combination score function is constructed to calculate defect score Pscore(m)：Wherein, i, j represent respectively the abscissa of pixel in the picture with Ordinate, A (m) represent m-th of detection window in Local neural network model SSD, and A (m) .conf represent Local neural network The score of m-th of testing result in model SSD, S represent the size of A (m) window, and B (i, j) represents coordinate as (i, j) Pixel value of the pixel in heatmap.
  10. 10. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (4), threshold value T is compared in setting, is realized by comparing the defects of calculating score with threshold size； If the defects of calculating score is less than given threshold T, then it is assumed that the testing result is false retrieval, is directly filtered out；If lack Fall into score and be more than given threshold T, then retain the defect area.
  11. 11. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (5), fabric defect region subgraph is divided into K super pictures using SLIC super-pixel segmentations algorithm Plain block, and using the heatmap figure extractions priori foreground point obtained in global neural network model, according to priori foreground point and Region contrast, regional space relation structure super-pixel significance function gamma (Pi)：
    Wherein, i1, j1Super-pixel node serial number is represented respectively,For i-th1Individual super-pixel node,For super-pixel node The normalization Euclidean distance of super-pixel node where to priori foreground point；For super-pixel nodeWith super-pixel section PointBetween normalization Euclidean distance；For super-pixel nodeCorresponding region Lab color spaces average value,For Super-pixel nodeCorresponding region Lab color spaces average value,Super-pixel region exists where priori foreground point The average value of Lab color spaces, K are super-pixel number, and α is regulatory factor.
  12. 12. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (5), during using heatmap figure extraction priori foreground points in global neural network model, root first Local heatmap figures are extracted in heatmap figures according to defecton image coordinate, then using N × N cover module in part Slided on heatmap, N is odd number, travels through the position that whole Zhang Suoshu parts heatmap images obtain maximum output value, maximum defeated Go out to be worth the priori prospect point coordinates of the i.e. defect in position；Cover module weight matrix isi2,j2Generation Coordinate of the table pixel in cover module, its specific formula are
    <mrow> <msub> <mi>&amp;omega;</mi> <mrow> <msub> <mi>i</mi> <mn>2</mn> </msub> <msub> <mi>j</mi> <mn>2</mn> </msub> </mrow> </msub> <mo>=</mo> <mi>u</mi> <mrow> <mo>(</mo> <mn>3</mn> <mo>-</mo> <msub> <mi>i</mi> <mn>2</mn> </msub> <mo>)</mo> </mrow> <mi>u</mi> <mrow> <mo>(</mo> <mn>3</mn> <mo>-</mo> <msub> <mi>j</mi> <mn>2</mn> </msub> <mo>)</mo> </mrow> <mo>+</mo> <mn>2</mn> <mo>&amp;times;</mo> <mi>u</mi> <mrow> <mo>(</mo> <msub> <mi>i</mi> <mn>2</mn> </msub> <mo>-</mo> <mn>3</mn> <mo>)</mo> </mrow> <mi>u</mi> <mrow> <mo>(</mo> <msub> <mi>j</mi> <mn>2</mn> </msub> <mo>-</mo> <mn>3</mn> <mo>)</mo> </mrow> <mo>&amp;times;</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>-</mo> <msup> <mrow> <mo>(</mo> <mrow> <mo>-</mo> <mn>1</mn> </mrow> <mo>)</mo> </mrow> <mrow> <msub> <mi>i</mi> <mn>2</mn> </msub> <mo>+</mo> <msub> <mi>j</mi> <mn>2</mn> </msub> </mrow> </msup> <mo>+</mo> <mn>4</mn> <mi>&amp;delta;</mi> <mo>(</mo> <mrow> <msub> <mi>i</mi> <mn>2</mn> </msub> <mo>-</mo> <mfrac> <mrow> <mi>N</mi> <mo>+</mo> <mn>1</mn> </mrow> <mn>2</mn> </mfrac> </mrow> <mo>)</mo> <mi>&amp;delta;</mi> <mo>(</mo> <mrow> <msub> <mi>j</mi> <mn>2</mn> </msub> <mo>-</mo> <mfrac> <mrow> <mi>N</mi> <mo>+</mo> <mn>1</mn> </mrow> <mn>2</mn> </mfrac> </mrow> <mo>)</mo> <mo>)</mo> </mrow> <mo>/</mo> <mn>2</mn> <msup> <mi>N</mi> <mn>2</mn> </msup> <mo>+</mo> <mn>14</mn> </mrow> ,
    Wherein δ () is impulse function, and u () is jump function, and N is template size.
  13. 13. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (5), the priori significance of each super-pixel point in textile image is calculated using significance function, And the pixel value using the priori saliency value as super-pixel point, build priori notable figure.
  14. 14. the fabric defect detection method according to claim 1 based on depth convolutional neural networks and vision significance, It is characterized in that：In step (6), priori notable figure is split using adaptive threshold OTSU algorithms, extracted in image Defect target, adaptive threshold choose formula and are：
    <mrow> <mi>T</mi> <mo>=</mo> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msub> <mi>&amp;alpha;</mi> <mn>1</mn> </msub> <msub> <mi>I</mi> <mrow> <mi>p</mi> <mi>r</mi> <mi>i</mi> <mi>o</mi> <mi>r</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>b</mi> <mn>1</mn> </msub> </mrow> </mtd> <mtd> <mrow> <mi>i</mi> <mi>f</mi> <mi> </mi> <msub> <mi>I</mi> <mrow> <mi>p</mi> <mi>r</mi> <mi>i</mi> <mi>o</mi> <mi>r</mi> </mrow> </msub> <mo>&gt;</mo> <mn>0.5</mn> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msub> <mi>&amp;alpha;</mi> <mn>2</mn> </msub> <msub> <mi>I</mi> <mrow> <mi>p</mi> <mi>r</mi> <mi>i</mi> <mi>o</mi> <mi>r</mi> </mrow> </msub> <mo>+</mo> <msub> <mi>b</mi> <mn>2</mn> </msub> </mrow> </mtd> <mtd> <mrow> <mi>e</mi> <mi>l</mi> <mi>s</mi> <mi>e</mi> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>,</mo> </mrow>
    Wherein α1, α2Respectively the threshold percentage factor one and the threshold percentage factor two, b1, b2Respectively threshold value shift factor one and threshold It is worth shift factor two, IpriorFor the pixel value of priori foreground point.
CN201711212830.0A 2017-11-28 2017-11-28 Fabric defect detection method based on deep convolutional neural network and visual saliency Active CN107833220B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201711212830.0A CN107833220B (en) 2017-11-28 2017-11-28 Fabric defect detection method based on deep convolutional neural network and visual saliency
PCT/CN2017/116837 WO2019104767A1 (en) 2017-11-28 2017-12-18 Fabric defect detection method based on deep convolutional neural network and visual saliency

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711212830.0A CN107833220B (en) 2017-11-28 2017-11-28 Fabric defect detection method based on deep convolutional neural network and visual saliency

Publications (2)

Publication Number Publication Date
CN107833220A true CN107833220A (en) 2018-03-23
CN107833220B CN107833220B (en) 2021-06-11

Family

ID=61646102

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711212830.0A Active CN107833220B (en) 2017-11-28 2017-11-28 Fabric defect detection method based on deep convolutional neural network and visual saliency

Country Status (2)

Country Link
CN (1) CN107833220B (en)
WO (1) WO2019104767A1 (en)

Cited By (56)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108562589A (en) * 2018-03-30 2018-09-21 慧泉智能科技（苏州）有限公司 A method of magnetic circuit material surface defect is detected
CN108664967A (en) * 2018-04-17 2018-10-16 上海交通大学 A kind of multimedia page vision significance prediction technique and system
CN108876781A (en) * 2018-06-26 2018-11-23 广东工业大学 Surface defect recognition method based on SSD algorithm
CN109033944A (en) * 2018-06-07 2018-12-18 西安电子科技大学 A kind of all-sky aurora image classification and crucial partial structurtes localization method and system
CN109239075A (en) * 2018-08-27 2019-01-18 北京百度网讯科技有限公司 Battery detection method and device
CN109239073A (en) * 2018-07-28 2019-01-18 西安交通大学 A kind of detection method of surface flaw for body of a motor car
CN109242830A (en) * 2018-08-18 2019-01-18 苏州翔升人工智能科技有限公司 A kind of machine vision technique detection method based on deep learning
CN109300109A (en) * 2018-08-23 2019-02-01 苏州富鑫林光电科技有限公司 The industrial intelligent image segmentation and identifying system and method for machine learning
CN109357679A (en) * 2018-11-16 2019-02-19 济南浪潮高新科技投资发展有限公司 A kind of indoor orientation method based on significant characteristics identification
CN109472769A (en) * 2018-09-26 2019-03-15 成都数之联科技有限公司 A kind of bad image defect detection method and system
CN109509172A (en) * 2018-09-25 2019-03-22 无锡动视宫原科技有限公司 A kind of liquid crystal display flaw detection method and system based on deep learning
CN109583450A (en) * 2018-11-27 2019-04-05 东南大学 Salient region detecting method based on feedforward neural network fusion vision attention priori
CN109740617A (en) * 2019-01-08 2019-05-10 国信优易数据有限公司 A kind of image detecting method and device
CN109813276A (en) * 2018-12-19 2019-05-28 五邑大学 A kind of antenna for base station has a down dip angle measuring method and its system
CN109872303A (en) * 2019-01-16 2019-06-11 北京交通大学 Surface defect visible detection method, device and electronic equipment
CN109961437A (en) * 2019-04-04 2019-07-02 江南大学 A kind of conspicuousness fabric defect detection method under the mode based on machine teaching
CN109978844A (en) * 2019-03-15 2019-07-05 维库（厦门）信息技术有限公司 One kind being based on the modular intelligent analysis method of deep learning and system
CN110033451A (en) * 2019-04-17 2019-07-19 国网山西省电力公司电力科学研究院 A kind of power components defect inspection method based on SSD framework
CN110111331A (en) * 2019-05-20 2019-08-09 中南大学 Honeycomb paper core defect inspection method based on machine vision
CN110223295A (en) * 2019-06-21 2019-09-10 安徽大学 Conspicuousness prediction technique and device based on deep neural network Color perception
CN110232406A (en) * 2019-05-28 2019-09-13 厦门大学 A kind of liquid crystal display panel CF image identification method based on statistical learning
CN110349146A (en) * 2019-07-11 2019-10-18 中原工学院 The building method of fabric defect identifying system based on lightweight convolutional neural networks
CN110472639A (en) * 2019-08-05 2019-11-19 山东工商学院 A kind of target extraction method based on conspicuousness prior information
CN110580701A (en) * 2019-08-13 2019-12-17 创新奇智(重庆)科技有限公司 Ready-made garment measurement and defect detection method based on computer vision
CN110619619A (en) * 2018-06-04 2019-12-27 杭州海康威视数字技术股份有限公司 Defect detection method and device and electronic equipment
CN110717880A (en) * 2018-07-11 2020-01-21 杭州海康威视数字技术股份有限公司 Defect detection method and device and electronic equipment
CN110766689A (en) * 2019-11-06 2020-02-07 深圳微品致远信息科技有限公司 Method and device for detecting article image defects based on convolutional neural network
CN110827263A (en) * 2019-11-06 2020-02-21 创新奇智(南京)科技有限公司 Magnetic shoe surface defect detection system and detection method based on visual identification technology
CN110889837A (en) * 2019-11-25 2020-03-17 东华大学 Cloth flaw detection method with flaw classification function
CN111062934A (en) * 2019-12-25 2020-04-24 陈金选 Real-time detection method for fabric image defects
CN111145163A (en) * 2019-12-30 2020-05-12 深圳市中钞科信金融科技有限公司 Paper wrinkle defect detection method and device
CN111210417A (en) * 2020-01-07 2020-05-29 创新奇智(北京)科技有限公司 Cloth defect detection method based on convolutional neural network
CN111275718A (en) * 2020-01-18 2020-06-12 江南大学 Clothes amount detection and color protection washing discrimination method based on significant region segmentation
CN111553265A (en) * 2020-04-27 2020-08-18 河北天元地理信息科技工程有限公司 Method and system for detecting internal defects of drainage pipeline
CN111563869A (en) * 2019-01-25 2020-08-21 宁波舜宇光电信息有限公司 Stain testing method for quality inspection of camera module
CN111582447A (en) * 2020-04-30 2020-08-25 电子科技大学 Closed loop detection method based on multiple network characteristics
CN111696092A (en) * 2020-06-11 2020-09-22 深圳市华汉伟业科技有限公司 Defect detection method and system based on feature comparison and storage medium
CN111929327A (en) * 2020-09-09 2020-11-13 深兰人工智能芯片研究院（江苏）有限公司 Cloth defect detection method and device
CN112102252A (en) * 2020-08-21 2020-12-18 北京无线电测量研究所 Method and device for detecting appearance defects of micro-strip antenna welding spot
CN112308860A (en) * 2020-10-28 2021-02-02 西北工业大学 Earth observation image semantic segmentation method based on self-supervision learning
CN112529829A (en) * 2019-08-28 2021-03-19 银河水滴科技（北京）有限公司 Training method and device for burr positioning and burr detection model
CN112561892A (en) * 2020-12-22 2021-03-26 东华大学 Defect detection method for printed and jacquard fabric
CN112598646A (en) * 2020-12-23 2021-04-02 山东产研鲲云人工智能研究院有限公司 Capacitance defect detection method and device, electronic equipment and storage medium
CN112767339A (en) * 2021-01-13 2021-05-07 哈尔滨工业大学 Surface defect detection method based on visual attention model
CN113298809A (en) * 2021-06-25 2021-08-24 成都飞机工业（集团）有限责任公司 Composite material ultrasonic image defect detection method based on deep learning and superpixel segmentation
CN113313706A (en) * 2021-06-28 2021-08-27 安徽南瑞继远电网技术有限公司 Power equipment defect image detection method based on detection reference point offset analysis
CN113379729A (en) * 2021-07-02 2021-09-10 四川启睿克科技有限公司 Image tiny anomaly detection method and device and computer readable storage medium
CN113506295A (en) * 2021-09-10 2021-10-15 启东市海信机械有限公司 Strip steel surface hot rolling slip defect detection method based on deep learning
CN113538503A (en) * 2021-08-21 2021-10-22 西北工业大学 Solar panel defect detection method based on infrared image
CN114463258A (en) * 2021-12-23 2022-05-10 中铁第一勘察设计院集团有限公司 Locomotive brake pad abrasion detection method based on deep learning technology
CN114565607A (en) * 2022-04-01 2022-05-31 南通沐沐兴晨纺织品有限公司 Fabric defect image segmentation method based on neural network
TWI769603B (en) * 2019-12-31 2022-07-01 大陸商鄭州富聯智能工坊有限公司 Image processing method and computer readable medium thereof
WO2023061173A1 (en) * 2021-10-13 2023-04-20 北京字节跳动网络技术有限公司 Image processing method and apparatus
CN116152807A (en) * 2023-04-14 2023-05-23 广东工业大学 Industrial defect semantic segmentation method based on U-Net network and storage medium
CN116402828A (en) * 2023-06-09 2023-07-07 江苏森标科技有限公司 Battery piece defect detection method and system based on saliency map
CN114463258B (en) * 2021-12-23 2024-04-30 中铁第一勘察设计院集团有限公司 Rolling stock brake pad abrasion detection method based on deep learning technology

Families Citing this family (100)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110632094B (en) * 2019-07-24 2022-04-19 北京中科慧眼科技有限公司 Pattern quality detection method, device and system based on point-by-point comparison analysis
CN110363254B (en) * 2019-08-12 2024-02-02 河北工业大学 Quick identification device for types of parts of flexible crankshaft production line
CN112417931B (en) * 2019-08-23 2024-01-26 河海大学常州校区 Method for detecting and classifying water surface objects based on visual saliency
CN110555831B (en) * 2019-08-29 2023-09-26 天津大学 Deep learning-based drainage pipeline defect segmentation method
US11727522B2 (en) * 2019-10-11 2023-08-15 Cargo Spectre Method, system, and apparatus for damage assessment and classification
CN110827260B (en) * 2019-11-04 2023-04-21 燕山大学 Cloth defect classification method based on LBP characteristics and convolutional neural network
CN111062910B (en) * 2019-11-13 2023-06-20 易思维（杭州）科技有限公司 Local threshold segmentation method and defect detection method
CN110910363A (en) * 2019-11-15 2020-03-24 上海交通大学 Insufficient solder joint detection method, system and medium based on machine vision and deep learning
CN111028207B (en) * 2019-11-22 2023-06-09 东华大学 Button flaw detection method based on instant-universal feature extraction network
CN110992329B (en) * 2019-11-28 2023-06-30 上海微创卜算子医疗科技有限公司 Product surface defect detection method, electronic equipment and readable storage medium
CN110969606B (en) * 2019-11-29 2023-08-08 华中科技大学 Texture surface defect detection method and system
CN111179278B (en) * 2019-12-16 2021-02-05 精英数智科技股份有限公司 Image detection method, device, equipment and storage medium
CN111275679B (en) * 2020-01-17 2023-06-02 同济大学 Image-based solar cell defect detection system and method
CN111242238B (en) * 2020-01-21 2023-12-26 北京交通大学 RGB-D image saliency target acquisition method
CN111402197B (en) * 2020-02-09 2023-06-16 西安工程大学 Detection method for colored fabric cut-parts defect area
CN111402203B (en) * 2020-02-24 2024-03-01 杭州电子科技大学 Fabric surface defect detection method based on convolutional neural network
CN111354031B (en) * 2020-03-16 2023-08-29 浙江一木智能科技有限公司 3D vision guidance system based on deep learning
CN111476766B (en) * 2020-03-31 2023-08-22 哈尔滨商业大学 Lung nodule CT image detection system based on deep learning
CN111640120B (en) * 2020-04-09 2023-08-29 之江实验室 Pancreas CT automatic segmentation method based on significance dense connection expansion convolution network
CN111598846B (en) * 2020-04-26 2024-01-05 鲲鹏通讯(昆山)有限公司 Method for detecting rail defects in tunnel based on YOLO
CN111652852B (en) * 2020-05-08 2024-03-29 浙江华睿科技股份有限公司 Product surface defect detection method, device and equipment
CN111582214B (en) * 2020-05-15 2023-05-12 中国科学院自动化研究所 Method, system and device for analyzing behavior of cage animal based on twin network
CN111768365B (en) * 2020-05-20 2023-05-30 太原科技大学 Solar cell defect detection method based on convolution neural network multi-feature fusion
CN113706437B (en) * 2020-05-21 2024-03-15 国网智能科技股份有限公司 Method and system for diagnosing defects of fine-granularity bolts of power transmission line
CN111860579A (en) * 2020-06-09 2020-10-30 广州寻星网络科技有限公司 Cloth cover flaw identification method and system and storage medium
CN111768386B (en) * 2020-06-30 2024-02-20 北京百度网讯科技有限公司 Product defect detection method, device, electronic equipment and storage medium
CN111899227A (en) * 2020-07-06 2020-11-06 北京交通大学 Automatic railway fastener defect acquisition and identification method based on unmanned aerial vehicle operation
CN112070851B (en) * 2020-07-07 2023-07-11 辽宁师范大学 Index map prediction method based on genetic algorithm and BP neural network
CN111950362B (en) * 2020-07-07 2024-04-16 西北大学 Golden monkey face image recognition method, device, equipment and storage medium
KR20220013071A (en) * 2020-07-24 2022-02-04 에스케이하이닉스 주식회사 Device for generating a depth map
CN111968076A (en) * 2020-07-24 2020-11-20 西安工程大学 Real-time fabric defect detection method based on S-YOLOV3
CN111882537B (en) * 2020-07-28 2023-12-15 研祥智能科技股份有限公司 Visual detection method and system
CN111860465A (en) * 2020-08-10 2020-10-30 华侨大学 Remote sensing image extraction method, device, equipment and storage medium based on super pixels
CN112085037B (en) * 2020-09-21 2024-04-09 国网吉林省电力有限公司电力科学研究院 Infrared thermal fault feature extraction and digital expression method for power transformation equipment
CN112270722A (en) * 2020-10-26 2021-01-26 西安工程大学 Digital printing fabric defect detection method based on deep neural network
CN112541884A (en) * 2020-10-30 2021-03-23 浙江大华技术股份有限公司 Defect detection method and apparatus, and computer-readable storage medium
CN112418019A (en) * 2020-11-08 2021-02-26 国家电网有限公司 Aerial power communication optical cable inspection system and method
CN112651972A (en) * 2020-11-11 2021-04-13 北京平恒智能科技有限公司 Positioning method using integral constraint of double positioning
CN112365478A (en) * 2020-11-13 2021-02-12 上海海事大学 Motor commutator surface defect detection model based on semantic segmentation
CN112508883A (en) * 2020-11-24 2021-03-16 南京航空航天大学 Self-adaptive defect identification method for train wheel tread taper hole detection
CN112488125B (en) * 2020-11-28 2021-12-14 重庆邮电大学 Reconstruction method and system based on high-speed visual diagnosis and BP neural network
CN112465815B (en) * 2020-12-17 2023-09-19 杭州电子科技大学 Remote sensing target significance detection method based on edge main body fusion information
CN112541912B (en) * 2020-12-23 2024-03-12 中国矿业大学 Rapid detection method and device for salient targets in mine sudden disaster scene
CN112785568A (en) * 2021-01-18 2021-05-11 厦门大学嘉庚学院 Magnetic shoe defect segmentation method based on deep learning
CN112907519A (en) * 2021-01-29 2021-06-04 广州信邦智能装备股份有限公司 Metal curved surface defect analysis system and method based on deep learning
CN112950547B (en) * 2021-02-03 2024-02-13 佛山科学技术学院 Machine vision detection method for lithium battery diaphragm defects based on deep learning
CN112991271B (en) * 2021-02-08 2024-02-02 西安理工大学 Aluminum profile surface defect visual detection method based on improved yolov3
CN112837311A (en) * 2021-03-02 2021-05-25 苏州零样本智能科技有限公司 Polyethylene particle defect detection and identification system and method based on deep learning
CN113033635A (en) * 2021-03-12 2021-06-25 中钞长城金融设备控股有限公司 Coin invisible image-text detection method and device
CN113269191A (en) * 2021-04-19 2021-08-17 内蒙古智诚物联股份有限公司 Crop leaf disease identification method and device and storage medium
CN113077454A (en) * 2021-04-19 2021-07-06 凌云光技术股份有限公司 Image defect fitting method, system and storage medium
CN113192018B (en) * 2021-04-23 2023-11-24 北京化工大学 Water-cooled wall surface defect video identification method based on fast segmentation convolutional neural network
CN113362276A (en) * 2021-04-26 2021-09-07 广东大自然家居科技研究有限公司 Visual detection method and system for plate
CN113313109A (en) * 2021-05-13 2021-08-27 中国计量大学 Semi-quantitative analysis method of fluorescence immunochromatographic test paper
CN113344857B (en) * 2021-05-13 2022-05-03 深圳市华汉伟业科技有限公司 Defect detection network training method, defect detection method and storage medium
CN113375676B (en) * 2021-05-26 2024-02-20 南京航空航天大学 Detector landing site positioning method based on impulse neural network
CN113379765B (en) * 2021-06-09 2023-09-08 大连海事大学 Road surface repair area extraction and repair type judgment method
CN113469224A (en) * 2021-06-16 2021-10-01 浙江大学 Rice classification method based on fusion of convolutional neural network and feature description operator
CN113470045B (en) * 2021-06-16 2024-04-16 浙江工业大学 Oral cavity CBCT image segmentation method based on super-pixel statistical characteristics and graph annotating force network
CN113554604B (en) * 2021-07-01 2024-02-02 常州大学 Melt-blown cloth defect area detection method based on machine vision
CN113554605B (en) * 2021-07-07 2023-09-26 常州大学 Fabric abnormal region detection method based on feature matching
CN113777030A (en) * 2021-07-08 2021-12-10 杭州信畅信息科技有限公司 Cloth surface defect detection device and method based on machine vision
CN113362347B (en) * 2021-07-15 2023-05-26 广东工业大学 Image defect region segmentation method and system based on super-pixel feature enhancement
CN113658108A (en) * 2021-07-22 2021-11-16 西南财经大学 Glass defect detection method based on deep learning
CN113657238B (en) * 2021-08-11 2024-02-02 南京精益安防系统科技有限公司 Fire early warning method based on neural network, storage medium and terminal equipment
CN113643275B (en) * 2021-08-29 2024-03-01 浙江工业大学 Ultrasonic defect detection method based on unsupervised manifold segmentation
CN113610843B (en) * 2021-08-30 2023-10-24 合肥工业大学智能制造技术研究院 Real-time defect identification system and method for optical fiber braiding layer
CN113793314A (en) * 2021-09-13 2021-12-14 河南丹圣源农业开发有限公司 Pomegranate maturity identification equipment and use method
CN113642529B (en) * 2021-09-17 2023-10-24 湖南科技大学 Barrier size prediction system and method based on GA-BP neural network
CN113834816A (en) * 2021-09-30 2021-12-24 江西省通讯终端产业技术研究院有限公司 Machine vision-based photovoltaic cell defect online detection method and system
CN113870236B (en) * 2021-10-09 2024-04-05 西北工业大学 Composite material defect nondestructive inspection method based on deep learning algorithm
CN114140400B (en) * 2021-11-16 2024-04-12 湖北中烟工业有限责任公司 Method for detecting cigarette packet label defect based on RANSAC and CNN algorithm
CN114092745B (en) * 2021-11-26 2024-04-05 中国人民解放军空军工程大学 Automatic calculation method for infrared thermal imaging layering defect area of carbon fiber composite material
CN114170174B (en) * 2021-12-02 2024-01-23 沈阳工业大学 CLANet steel rail surface defect detection system and method based on RGB-D image
CN114972225B (en) * 2022-05-16 2023-04-25 上海可明科技有限公司 Deep learning-based two-stage photovoltaic panel defect detection method
CN114926436A (en) * 2022-05-20 2022-08-19 南通沐沐兴晨纺织品有限公司 Defect detection method for periodic pattern fabric
CN115578660B (en) * 2022-11-09 2023-04-07 牧马人(山东)勘察测绘集团有限公司 Land block segmentation method based on remote sensing image
CN115960605B (en) * 2022-12-09 2023-10-24 西南政法大学 Multicolor fluorescent carbon dot and application thereof
CN115690105B (en) * 2022-12-30 2023-03-21 无锡康贝电子设备有限公司 Milling cutter scratch detection method based on computer vision
CN115908142B (en) * 2023-01-06 2023-05-09 诺比侃人工智能科技(成都)股份有限公司 Visual identification-based damage inspection method for tiny contact net parts
CN116797533B (en) * 2023-03-24 2024-01-23 东莞市冠锦电子科技有限公司 Appearance defect detection method and system for power adapter
CN116030061B (en) * 2023-03-29 2023-08-22 深圳市捷超行模具有限公司 Silica gel molding effect detection method based on vision
CN116309589B (en) * 2023-05-22 2023-08-01 季华实验室 Sheet metal part surface defect detection method and device, electronic equipment and storage medium
CN116309600B (en) * 2023-05-24 2023-08-04 山东金佳成工程材料有限公司 Environment-friendly textile quality detection method based on image processing
CN116523901B (en) * 2023-06-20 2023-09-19 东莞市京品精密模具有限公司 Punching die detection method based on computer vision
CN116485801B (en) * 2023-06-26 2023-09-12 山东兰通机电有限公司 Rubber tube quality online detection method and system based on computer vision
CN116596922B (en) * 2023-07-17 2023-09-12 山东龙普太阳能股份有限公司 Production quality detection method of solar water heater
CN116630813B (en) * 2023-07-24 2023-09-26 青岛奥维特智能科技有限公司 Highway road surface construction quality intelligent detection system
CN116681992B (en) * 2023-07-29 2023-10-20 河南省新乡生态环境监测中心 Ammonia nitrogen detection method based on neural network
CN117315578A (en) * 2023-08-14 2023-12-29 金钱猫科技股份有限公司 Monitoring method and system for rust area expansion by combining classification network
CN116935179B (en) * 2023-09-14 2023-12-08 海信集团控股股份有限公司 Target detection method and device, electronic equipment and storage medium
CN116934749B (en) * 2023-09-15 2023-12-19 山东虹纬纺织有限公司 Textile flaw rapid detection method based on image characteristics
CN117197006A (en) * 2023-09-27 2023-12-08 上海世禹精密设备股份有限公司 Region filtering method, region filtering device, electronic equipment and computer readable storage medium
CN116993745B (en) * 2023-09-28 2023-12-19 山东辉瑞管业有限公司 Method for detecting surface leakage of water supply pipe based on image processing
CN117437188A (en) * 2023-10-17 2024-01-23 广东电力交易中心有限责任公司 Insulator defect detection system for smart power grid
CN117132736B (en) * 2023-10-25 2024-02-13 深圳市广通软件有限公司 Stadium modeling method and system based on meta universe
CN117152148B (en) * 2023-10-31 2023-12-29 南通杰元纺织品有限公司 Method for detecting defect of wool spots of textile
CN117314901B (en) * 2023-11-28 2024-02-02 闽都创新实验室 Scale-adaptive chip detection neural network system
CN117351062B (en) * 2023-12-04 2024-02-23 尚特杰电力科技有限公司 Fan blade defect diagnosis method, device and system and electronic equipment
CN117541587B (en) * 2024-01-10 2024-04-02 山东建筑大学 Solar panel defect detection method, system, electronic equipment and storage medium

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB201217395D0 (en) * 2012-09-28 2012-11-14 Omg Plc Determination of position from images and associated camera positions
EP2570970A1 (en) * 2011-09-16 2013-03-20 Technische Universität Berlin Method and system for the automatic analysis of an image of a biological sample
CN103729842A (en) * 2013-12-20 2014-04-16 中原工学院 Fabric defect detection method based on local statistical characteristics and overall significance analysis
CN105701508A (en) * 2016-01-12 2016-06-22 西安交通大学 Global-local optimization model based on multistage convolution neural network and significant detection algorithm
CN105701477A (en) * 2016-02-19 2016-06-22 中原工学院 Fabric defect detection method based on steady wavelet transform visual saliency
CN106203430A (en) * 2016-07-07 2016-12-07 北京航空航天大学 A kind of significance object detecting method based on foreground focused degree and background priori
CN106447658A (en) * 2016-09-26 2017-02-22 西北工业大学 Significant target detection method based on FCN (fully convolutional network) and CNN (convolutional neural network)
CN106599830A (en) * 2016-12-09 2017-04-26 中国科学院自动化研究所 Method and apparatus for positioning face key points
CN106778634A (en) * 2016-12-19 2017-05-31 江苏慧眼数据科技股份有限公司 A kind of conspicuousness human region detection method based on region fusion
CN107169954A (en) * 2017-04-18 2017-09-15 华南理工大学 A kind of image significance detection method based on parallel-convolution neutral net
CN107274419A (en) * 2017-07-10 2017-10-20 北京工业大学 A kind of deep learning conspicuousness detection method based on global priori and local context
CN107292875A (en) * 2017-06-29 2017-10-24 西安建筑科技大学 A kind of conspicuousness detection method based on global Local Feature Fusion

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8660368B2 (en) * 2011-03-16 2014-02-25 International Business Machines Corporation Anomalous pattern discovery

Patent Citations (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2570970A1 (en) * 2011-09-16 2013-03-20 Technische Universität Berlin Method and system for the automatic analysis of an image of a biological sample
GB201217395D0 (en) * 2012-09-28 2012-11-14 Omg Plc Determination of position from images and associated camera positions
CN103729842A (en) * 2013-12-20 2014-04-16 中原工学院 Fabric defect detection method based on local statistical characteristics and overall significance analysis
CN105701508A (en) * 2016-01-12 2016-06-22 西安交通大学 Global-local optimization model based on multistage convolution neural network and significant detection algorithm
CN105701477A (en) * 2016-02-19 2016-06-22 中原工学院 Fabric defect detection method based on steady wavelet transform visual saliency
CN106203430B (en) * 2016-07-07 2017-11-03 北京航空航天大学 A kind of conspicuousness object detecting method based on foreground focused degree and background priori
CN106203430A (en) * 2016-07-07 2016-12-07 北京航空航天大学 A kind of significance object detecting method based on foreground focused degree and background priori
CN106447658A (en) * 2016-09-26 2017-02-22 西北工业大学 Significant target detection method based on FCN (fully convolutional network) and CNN (convolutional neural network)
CN106599830A (en) * 2016-12-09 2017-04-26 中国科学院自动化研究所 Method and apparatus for positioning face key points
CN106778634A (en) * 2016-12-19 2017-05-31 江苏慧眼数据科技股份有限公司 A kind of conspicuousness human region detection method based on region fusion
CN107169954A (en) * 2017-04-18 2017-09-15 华南理工大学 A kind of image significance detection method based on parallel-convolution neutral net
CN107292875A (en) * 2017-06-29 2017-10-24 西安建筑科技大学 A kind of conspicuousness detection method based on global Local Feature Fusion
CN107274419A (en) * 2017-07-10 2017-10-20 北京工业大学 A kind of deep learning conspicuousness detection method based on global priori and local context

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
ZHILIU等: "Superpixel-Based Spatiotemporal Saliency Detection", 《 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY》 *
张文杰: "基于图像配准与视觉显著性检测的指针仪表识别研究", 《中国博士学位论文全文数据库·信息科技辑》 *
张文杰等: "基于对比度与空间位置特征的显著性区域检测", 《吉林大学学报》 *
景军锋等: "应用深度卷积神经网络的色织物缺陷检测", 《纺织学报》 *
马云鹏等: "金属表面缺陷自适应分割算法", 《仪器仪表学报》 *
魏娟: "基于机器视觉的坯布疵点检测方法研究", 《中国优秀硕士学位论文全文数据库·工程科技Ⅰ辑》 *

Cited By (80)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108562589A (en) * 2018-03-30 2018-09-21 慧泉智能科技（苏州）有限公司 A method of magnetic circuit material surface defect is detected
CN108562589B (en) * 2018-03-30 2020-12-01 慧泉智能科技（苏州）有限公司 Method for detecting surface defects of magnetic circuit material
CN108664967B (en) * 2018-04-17 2020-08-25 上海媒智科技有限公司 Method and system for predicting visual saliency of multimedia page
CN108664967A (en) * 2018-04-17 2018-10-16 上海交通大学 A kind of multimedia page vision significance prediction technique and system
CN110619619A (en) * 2018-06-04 2019-12-27 杭州海康威视数字技术股份有限公司 Defect detection method and device and electronic equipment
CN109033944A (en) * 2018-06-07 2018-12-18 西安电子科技大学 A kind of all-sky aurora image classification and crucial partial structurtes localization method and system
CN109033944B (en) * 2018-06-07 2021-09-24 西安电子科技大学 Method and system for classifying all-sky aurora images and positioning key local structure
CN108876781A (en) * 2018-06-26 2018-11-23 广东工业大学 Surface defect recognition method based on SSD algorithm
CN110717880A (en) * 2018-07-11 2020-01-21 杭州海康威视数字技术股份有限公司 Defect detection method and device and electronic equipment
CN109239073A (en) * 2018-07-28 2019-01-18 西安交通大学 A kind of detection method of surface flaw for body of a motor car
CN109242830A (en) * 2018-08-18 2019-01-18 苏州翔升人工智能科技有限公司 A kind of machine vision technique detection method based on deep learning
CN109300109A (en) * 2018-08-23 2019-02-01 苏州富鑫林光电科技有限公司 The industrial intelligent image segmentation and identifying system and method for machine learning
CN109300109B (en) * 2018-08-23 2020-10-09 苏州富鑫林光电科技有限公司 Machine learning industrial intelligent image segmentation and recognition system
CN109239075A (en) * 2018-08-27 2019-01-18 北京百度网讯科技有限公司 Battery detection method and device
CN109509172A (en) * 2018-09-25 2019-03-22 无锡动视宫原科技有限公司 A kind of liquid crystal display flaw detection method and system based on deep learning
CN109472769A (en) * 2018-09-26 2019-03-15 成都数之联科技有限公司 A kind of bad image defect detection method and system
CN109357679B (en) * 2018-11-16 2022-04-19 山东浪潮科学研究院有限公司 Indoor positioning method based on significance characteristic recognition
CN109357679A (en) * 2018-11-16 2019-02-19 济南浪潮高新科技投资发展有限公司 A kind of indoor orientation method based on significant characteristics identification
CN109583450A (en) * 2018-11-27 2019-04-05 东南大学 Salient region detecting method based on feedforward neural network fusion vision attention priori
CN109813276B (en) * 2018-12-19 2021-01-26 五邑大学 Base station antenna downward inclination angle measuring method and system
CN109813276A (en) * 2018-12-19 2019-05-28 五邑大学 A kind of antenna for base station has a down dip angle measuring method and its system
CN109740617A (en) * 2019-01-08 2019-05-10 国信优易数据有限公司 A kind of image detecting method and device
CN109872303A (en) * 2019-01-16 2019-06-11 北京交通大学 Surface defect visible detection method, device and electronic equipment
CN111563869B (en) * 2019-01-25 2023-07-21 宁波舜宇光电信息有限公司 Stain test method for quality inspection of camera module
CN111563869A (en) * 2019-01-25 2020-08-21 宁波舜宇光电信息有限公司 Stain testing method for quality inspection of camera module
CN109978844A (en) * 2019-03-15 2019-07-05 维库（厦门）信息技术有限公司 One kind being based on the modular intelligent analysis method of deep learning and system
CN109961437A (en) * 2019-04-04 2019-07-02 江南大学 A kind of conspicuousness fabric defect detection method under the mode based on machine teaching
CN110033451A (en) * 2019-04-17 2019-07-19 国网山西省电力公司电力科学研究院 A kind of power components defect inspection method based on SSD framework
CN110111331A (en) * 2019-05-20 2019-08-09 中南大学 Honeycomb paper core defect inspection method based on machine vision
CN110232406A (en) * 2019-05-28 2019-09-13 厦门大学 A kind of liquid crystal display panel CF image identification method based on statistical learning
CN110223295B (en) * 2019-06-21 2022-05-03 安徽大学 Significance prediction method and device based on deep neural network color perception
CN110223295A (en) * 2019-06-21 2019-09-10 安徽大学 Conspicuousness prediction technique and device based on deep neural network Color perception
CN110349146A (en) * 2019-07-11 2019-10-18 中原工学院 The building method of fabric defect identifying system based on lightweight convolutional neural networks
CN110472639A (en) * 2019-08-05 2019-11-19 山东工商学院 A kind of target extraction method based on conspicuousness prior information
CN110472639B (en) * 2019-08-05 2023-04-18 山东工商学院 Target extraction method based on significance prior information
CN110580701A (en) * 2019-08-13 2019-12-17 创新奇智(重庆)科技有限公司 Ready-made garment measurement and defect detection method based on computer vision
CN112529829A (en) * 2019-08-28 2021-03-19 银河水滴科技（北京）有限公司 Training method and device for burr positioning and burr detection model
CN112529829B (en) * 2019-08-28 2024-04-19 银河水滴科技（北京）有限公司 Training method and device for burr positioning and burr detection model
CN110827263A (en) * 2019-11-06 2020-02-21 创新奇智(南京)科技有限公司 Magnetic shoe surface defect detection system and detection method based on visual identification technology
CN110766689A (en) * 2019-11-06 2020-02-07 深圳微品致远信息科技有限公司 Method and device for detecting article image defects based on convolutional neural network
CN110889837A (en) * 2019-11-25 2020-03-17 东华大学 Cloth flaw detection method with flaw classification function
CN111062934B (en) * 2019-12-25 2023-10-13 陈金选 Fabric image defect real-time detection method
CN111062934A (en) * 2019-12-25 2020-04-24 陈金选 Real-time detection method for fabric image defects
CN111145163A (en) * 2019-12-30 2020-05-12 深圳市中钞科信金融科技有限公司 Paper wrinkle defect detection method and device
TWI769603B (en) * 2019-12-31 2022-07-01 大陸商鄭州富聯智能工坊有限公司 Image processing method and computer readable medium thereof
CN111210417B (en) * 2020-01-07 2023-04-07 创新奇智(北京)科技有限公司 Cloth defect detection method based on convolutional neural network
CN111210417A (en) * 2020-01-07 2020-05-29 创新奇智(北京)科技有限公司 Cloth defect detection method based on convolutional neural network
CN111275718B (en) * 2020-01-18 2024-01-30 江南大学 Clothes amount detection and color protection washing discrimination method based on significant region segmentation
CN111275718A (en) * 2020-01-18 2020-06-12 江南大学 Clothes amount detection and color protection washing discrimination method based on significant region segmentation
CN111553265A (en) * 2020-04-27 2020-08-18 河北天元地理信息科技工程有限公司 Method and system for detecting internal defects of drainage pipeline
CN111553265B (en) * 2020-04-27 2021-10-29 河北天元地理信息科技工程有限公司 Method and system for detecting internal defects of drainage pipeline
CN111582447A (en) * 2020-04-30 2020-08-25 电子科技大学 Closed loop detection method based on multiple network characteristics
CN111696092A (en) * 2020-06-11 2020-09-22 深圳市华汉伟业科技有限公司 Defect detection method and system based on feature comparison and storage medium
CN111696092B (en) * 2020-06-11 2023-08-25 深圳市华汉伟业科技有限公司 Defect detection method and system based on feature comparison and storage medium
CN112102252A (en) * 2020-08-21 2020-12-18 北京无线电测量研究所 Method and device for detecting appearance defects of micro-strip antenna welding spot
CN112102252B (en) * 2020-08-21 2023-11-28 北京无线电测量研究所 Method and device for detecting appearance defects of welding spots of microstrip antenna
CN111929327A (en) * 2020-09-09 2020-11-13 深兰人工智能芯片研究院（江苏）有限公司 Cloth defect detection method and device
CN112308860A (en) * 2020-10-28 2021-02-02 西北工业大学 Earth observation image semantic segmentation method based on self-supervision learning
CN112308860B (en) * 2020-10-28 2024-01-12 西北工业大学 Earth observation image semantic segmentation method based on self-supervision learning
CN112561892A (en) * 2020-12-22 2021-03-26 东华大学 Defect detection method for printed and jacquard fabric
CN112598646A (en) * 2020-12-23 2021-04-02 山东产研鲲云人工智能研究院有限公司 Capacitance defect detection method and device, electronic equipment and storage medium
CN112767339A (en) * 2021-01-13 2021-05-07 哈尔滨工业大学 Surface defect detection method based on visual attention model
CN112767339B (en) * 2021-01-13 2023-12-29 哈尔滨工业大学 Surface defect detection method based on visual attention model
CN113298809A (en) * 2021-06-25 2021-08-24 成都飞机工业（集团）有限责任公司 Composite material ultrasonic image defect detection method based on deep learning and superpixel segmentation
CN113298809B (en) * 2021-06-25 2022-04-08 成都飞机工业（集团）有限责任公司 Composite material ultrasonic image defect detection method based on deep learning and superpixel segmentation
CN113313706A (en) * 2021-06-28 2021-08-27 安徽南瑞继远电网技术有限公司 Power equipment defect image detection method based on detection reference point offset analysis
CN113379729B (en) * 2021-07-02 2023-07-25 四川启睿克科技有限公司 Image tiny abnormality detection method, device and computer readable storage medium
CN113379729A (en) * 2021-07-02 2021-09-10 四川启睿克科技有限公司 Image tiny anomaly detection method and device and computer readable storage medium
CN113538503A (en) * 2021-08-21 2021-10-22 西北工业大学 Solar panel defect detection method based on infrared image
CN113538503B (en) * 2021-08-21 2023-09-01 西北工业大学 Solar panel defect detection method based on infrared image
CN113506295B (en) * 2021-09-10 2021-11-26 启东市海信机械有限公司 Strip steel surface hot rolling slip defect detection method based on deep learning
CN113506295A (en) * 2021-09-10 2021-10-15 启东市海信机械有限公司 Strip steel surface hot rolling slip defect detection method based on deep learning
WO2023061173A1 (en) * 2021-10-13 2023-04-20 北京字节跳动网络技术有限公司 Image processing method and apparatus
CN114463258A (en) * 2021-12-23 2022-05-10 中铁第一勘察设计院集团有限公司 Locomotive brake pad abrasion detection method based on deep learning technology
CN114463258B (en) * 2021-12-23 2024-04-30 中铁第一勘察设计院集团有限公司 Rolling stock brake pad abrasion detection method based on deep learning technology
CN114565607A (en) * 2022-04-01 2022-05-31 南通沐沐兴晨纺织品有限公司 Fabric defect image segmentation method based on neural network
CN116152807B (en) * 2023-04-14 2023-09-05 广东工业大学 Industrial defect semantic segmentation method based on U-Net network and storage medium
CN116152807A (en) * 2023-04-14 2023-05-23 广东工业大学 Industrial defect semantic segmentation method based on U-Net network and storage medium
CN116402828B (en) * 2023-06-09 2023-09-08 江苏森标科技有限公司 Battery piece defect detection method and system based on saliency map
CN116402828A (en) * 2023-06-09 2023-07-07 江苏森标科技有限公司 Battery piece defect detection method and system based on saliency map

Also Published As

Publication number Publication date
CN107833220B (en) 2021-06-11
WO2019104767A1 (en) 2019-06-06

Similar Documents

Publication Publication Date Title
CN107833220A (en) Fabric defect detection method based on depth convolutional neural networks and vision significance
CN111723860B (en) Target detection method and device
CN109961049B (en) Cigarette brand identification method under complex scene
Zhang et al. Fabric defect detection using salience metric for color dissimilarity and positional aggregation
CN108765465B (en) Unsupervised SAR image change detection method
CN112818862B (en) Face tampering detection method and system based on multi-source clues and mixed attention
CN110852316B (en) Image tampering detection and positioning method adopting convolution network with dense structure
CN110060237A (en) A kind of fault detection method, device, equipment and system
CN106875395B (en) Super-pixel-level SAR image change detection method based on deep neural network
CN110119728A (en) Remote sensing images cloud detection method of optic based on Multiscale Fusion semantic segmentation network
CN107563999A (en) A kind of chip defect recognition methods based on convolutional neural networks
CN109711288A (en) Remote sensing ship detecting method based on feature pyramid and distance restraint FCN
CN107220649A (en) A kind of plain color cloth defects detection and sorting technique
CN109613006A (en) A kind of fabric defect detection method based on end-to-end neural network
CN106780485A (en) SAR image change detection based on super-pixel segmentation and feature learning
CN109087330A (en) It is a kind of based on by slightly to the moving target detecting method of smart image segmentation
CN103208097A (en) Principal component analysis collaborative filtering method for image multi-direction morphological structure grouping
CN110008900A (en) A kind of visible remote sensing image candidate target extracting method by region to target
CN114742758A (en) Cell nucleus classification method in full-field digital slice histopathology picture
CN110458812A (en) A kind of similar round fruit defects detection method based on color description and sparse expression
CN111738310B (en) Material classification method, device, electronic equipment and storage medium
CN113609944A (en) Silent in-vivo detection method
CN108932471A (en) A kind of vehicle checking method
CN115471646A (en) Unsupervised rapid image anomaly detection and positioning method
CN111461139B (en) Multi-target visual saliency layered detection method in complex scene

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant